Corticosteroid excess is associated with changes in memory and hippocampal structure. A consistent finding during corticosteroid exposure or following stress is a decline in performance on declarative memory tasks (e.g., word lists, paragraph recall). Impairment in declarative memory and hippocampal atrophy have been reported in patients with corticosteroid excess due to Cushing's disease, and, by our group, in medically ill patients receiving prescription corticosteroid therapy. These findings have important implications for patients with mood disorders, as a subset of people with major depressive disorder and bipolar disorder show evidence of HPA axis activation and elevated cortisol. In animal models, hippocampal changes secondary to corticosteroids can be attenuated with agents that modulate excitatory amino acids. Histological changes in animals can be prevented with glutamate release inhibitors or N-methyl-D-aspartate (NMDA) receptor antagonists. Over the past 15 years, our group has developed a research program using patients in medical settings receiving prescription corticosteroid (e.g., prednisone) therapy as a model system to explore the effects of cortisol elevations on the human brain. This is also a large and clinically important patient population that frequently has memory impairment as well as mood symptoms related to corticosteroid therapy. We have reported depressive symptomatology, deficits in declarative memory, changes in N-acetyl aspartate (NAA, a marker of neuronal viability), and reduction in hippocampal volume during chronic exposure to prednisone. We conducted a placebo-controlled proof-of-concept study of the NMDA receptor antagonist memantine in corticosteroid-treated patients and found statistically significant improvement in declarative memory. A larger and longer definitive trial of memantine is now proposed using neuroimaging as well as memory as outcome measures. We propose a randomized, double-blind, placebo-controlled crossover trial of memantine in 50 outpatients with asthma receiving chronic oral chronic corticosteroid therapy. We hypothesize that the group receiving memantine will show improvement in declarative memory (primary outcome measure) relative to the placebo group. We will also obtain structural MRI and HMRS to examine changes in hippocampal and amygdalar volumes and levels of NAA and glutamate. We have assembled a research team with expertise in mood disorders, neuroendocrinology, clinical trials, neuropsychology, asthma, and statistics to conduct the study. The findings will have implications for patients with mood disorders, substance use disorders, dementias, Cushing's disease, and the millions treated each year with prescription corticosteroids.